ENDANGERED SPECIES RESEARCHEndang Species Res

Vol. 14: 259–270, 2011doi: 10.3354/esr00362

Published online September 29

INTRODUCTION

The island of Madagascar, a hotspot of biodiversity(Myers et al. 2000), has experienced anthropogenicmodification for nearly 2 millennia, but habitatdestruction has recently accelerated in intensity(Green & Sussman 1990, Casse et al. 2004). Northernand eastern rainforests have received vital researchand conservation attention, but much of the endemicspiny forest of southern Madagascar is poorly under-stood, and thus should be a focus of future researchefforts (Bodin et al. 2006). This arid eco-region, esti-mated to have been 5 million hectares large in 2003

(Wells 2003), is highly impacted by continual anthro-pogenic change, including slash and burn agricul-ture, charcoal production and livestock maintenance(Scott et al. 2006). Spiny forests contain a high diver-sity of lemurs, including members of the taxonomicfamilies Cheirogaleidae, Lepilemuridae, Lemuridaeand Indridae (Sussman 2002). Thus, the spiny forestecoregion is a conservation focal point, and itspreservation is essential to sustaining lemur diversitythroughout the region.

A key species in the maintenance of spiny forests isthe ring-tailed lemur Lemur catta (Bodin et al. 2006,Scott et al. 2006). L. catta are highly adaptable and

© Inter-Research 2011 · www.int-res.com*Email: nicholas.ellwanger@utsa.edu

Variations in behavioural patterns betweenLemur catta groups living in different forest types:

implications for conservation

Nicholas Ellwanger1,*, Lisa Gould2

1Department of Anthropology, University of Texas at San Antonio, San Antonio, Texas 78249, USA2Department of Anthropology, University of Victoria, Victoria, British Columbia V8W 3P5, Canada

ABSTRACT: Implementing effective primate conservation strategies requires knowledge of spe-cies’ ecological and behavioural parameters. The protection of all habitat types required for pre-serving viable dispersion and gene flow between disparate populations necessitates a thoroughunderstanding of the behavioural flexibility of a species. Ring-tailed lemurs Lemur catta inhabit aplethora of habitats throughout southern Madagascar but, hitherto, research has been restrictedprimarily to riverine gallery forests. The present study presents comparative data on the activitybudgets and foraging behaviour of 2 L. catta groups living in different habitats at Berenty Reserve,Madagascar. Data were collected for 2 mo on groups living in a xerophytic spiny forest and a river-ine gallery forest. The spiny forest group spent significantly more time in inactive behaviours com-pared with the gallery forest group. This strategy of apparent energy minimization allows L. cattato thermoregulate despite decreased water availability and high ambient temperatures. The spinyforest group spent significantly more time feeding on small fruits of succulent trees, which weassociate with the group’s significantly higher rate of feeding agonism compared with the galleryforest group. These results demonstrate the high behavioural and ecological flexibility of L. catta,suggesting that xerophytic spiny forests might be a vital habitat type for L. catta conservationefforts. Future research should advance knowledge of the resources needed to sustain L. cattapopulations in other xerophytic spiny forests.

KEY WORDS: Lemur catta · Spiny forest · Gallery forest · Activity budget · Behavioural flexibility

Resale or republication not permitted without written consent of the publisher

OPENPEN ACCESSCCESS

currently range in numerous habitat types through-out southern Madagascar, including gallery forest,xerophytic forest, brush and scrub, spiny, dry decid-uous forest, sub-alpine ericoid bush and rocky out-crop forest surrounded by anthropogenic savannah(Sussman et al. 2003, Goodman et al. 2006, Gould2006). Although L. catta occur in these varied habi-tats, the majority of L. catta behavioural and popula-tion ecology field studies have been conducted inriverine gallery forests (e.g. Jolly 1966, Sussman1974, 1977, 1991, 1992, Gould 1990, 1992, 1996a,b,1997a,b, Sauther 1993, 1994, 1998, Nakamichi &Koyama 1997, Gould et al. 1999, 2003, Koyama et al.2001, Jolly et al. 2002, but see Gould et al. 2009). Lit-tle research has focused on populations in other habi-tats, although non-gallery forest research is begin-ning to occur at a few sites, including sub-alpineericoid bush (Goodman & Langrand 1996, Goodman& Rasolonandrasana 2001), spiny forest (Gould et al.2009, 2011, E. Kelley unpubl. data, M. LaFleurunpubl. data) and rocky outcrop forests surroundedby anthropogenic crop production (Gould et al. 2010,A. Cameron & L. Gould unpubl. data).

Goodman et al. (2006) suggest that Lemur cattamay have evolved in arid habitats rather thangallery forests throughout southern Madagascarand, although they are found in smaller populationsin these areas than in riverine gallery forests (Good-man & Langrand 1996, Sussman et al. 2003, Jolly etal. 2006), several physical and behavioural traitssupport this claim. L. catta is characterized by theability to live in areas without groundwater andwith few water-based food items (Goodman et al.2006). Furthermore, L. catta has the ability todecrease basal metabolic rate when food resourceavailability is low (Pereira et al. 1999). L. catta ismore terrestrial than other lemur taxa, spending anaverage of 30% of its time on the ground (Sussman1977). Moreover, L. catta populations can reboundfrom periods of decreased habitat productivity, nat-ural disasters and extended periods of drought, astheir relatively precocial life history pattern allowsjuveniles to reach reproductive age quickly, com-pared with many Haplorhine primates (Gould et al.1999, 2003, Gould 2006, Goodman et al. 2006), andfemales normally give birth once per year (Jolly1966, 1984, Sauther 1991). Thus, L. catta have beencharacterized as a ‘weed’ species because of theirabilities to live in harsh environments with extremesof heat and frost, exploit a variety of high and lowquality food items, and survive periods of intensewater scarcity (Gould et al. 1999, 2003, Goodman etal. 2006, Gould 2006).

Lemur catta is currently categorized as NearThreatened on the IUCN Red List, because one quar-ter of its population has disappeared over the past25 yr owing to substantial habitat loss (IUCN 2009).The future of this species and the endemic spinyforests are intricately tied together. Because individ-uals and groups disperse across dry riverbeds andopen habitats (Goodman et al. 2006), L. catta serve asprimary seed dispersers between different patches ofspiny forest habitats, a function that can maintainseemingly fragmented pockets of forest (Ganzhorn etal. 1999, Bodin et al. 2006). Although traditionaltaboos on vast forest destruction may help to main-tain, even unintentionally, L. catta populations at asustainable level (Bodin et al. 2006), small spiny for-est patches are highly vulnerable to anthropogenicpressures, decreasing the ability of L. catta to occupysuch forests. Intensive agricultural practices, specifi-cally tavy slash and burn forest clearance, havereduced viable habitats for L. catta throughout theirrange (Fenn & Rebara 2003, Kelley et al. 2007). Sus-tained spiny forest habitat destruction limits L. cattagene flow because of the lack of small forest patchesthrough which to disperse (Bodin et al. 2006). Thus,an improved effort to ensure the viability of L. cattapopulations must identify spiny forest fragments thatcan be effectively protected and monitored.

Long-term conservation strategies require knowl-edge of the resource use and behaviour of a specieswithin its geographic range. The lack of comparabledata on Lemur catta in highly diverse habitats makescurrent knowledge of this species’ behavioural ecol-ogy incomplete (Gould 2006). Furthermore, proposedmodels of L. catta evolution may be better under-stood with a more complete account of behaviouraltrends in habitats in which its evolutionary historyhas occurred (Gould 2006). The present study pro-vides initial data on behavioural strategies of L. cattain spiny forest habitats and aims to engender futurestudies that will enhance the ability to sustain andconnect genetically diverse populations of this spe-cies of lemur.

In this study, we contrast activity budgets, feedingbehaviour and feeding competition of Lemur cattaliving in spiny forest habitats with those living ingallery forest habitats in Berenty Reserve, southernMadagascar. This comparative study provides a bet-ter understanding of simultaneous behavioural vari-ation and reports behavioural strategies of L. cattain xerophytic spiny forests. We report differingbehavioural strategies used in ecological settingsthat vary in ambient temperature, water availability,canopy cover and food type during the annual birth

Endang Species Res 14: 259–270, 2011260

Ellwanger et al.: Behavioural flexibility in Lemur catta

and early lactation season (September to Novem-ber). This period coincides with the height of thedry season in southern Madagascar (Jolly 1984,Sauther 1994, 1998), which is characterized by veryhigh ambient temperatures and a decrease in avail-able drinking water (Fenn 2003). Because resourceabundance is low during this period compared withthe rainy season (Jolly 1984, Rasamimanana & Ra -fidinarivo 1993, Sauther 1998, Sauther et al. 1999),strict reproductive seasonality in this primate meansthat lactating females are in need of, and competingfor, scarce resources (Sauther 1993, Sauther et al.1999).

MATERIALS AND METHODS

Research site: Berenty Reserve, Madagascar

Berenty Private Reserve, southern Madagascar, is a240 km2 ecotourism site located on the banks of theMandare River: 25° 0.5’ S, 46° 18.5’ E (Jolly 1966,Jolly et al. 2006) (Fig. 1), owned and managed by thede Heaulme family (Jolly et al. 2006). The reservereceives 400 to 520 mm of rainfall between Novem-ber and February, comprising 70% of the annual

rainfall (Pride 2003). During our study, maximumdaily mid-day temperatures in the open spiny forestreached 40 to 45°C, whereas temperatures weregreater than 5°C cooler at the same time in the closedcanopy gallery forest.

Data were collected between 4 September and4 November 2006, corresponding to the end of thedry season in southern Madagascar and the late ges-tation and early to mid lactation seasons of Lemurcatta. Data were collected on L. catta groups in 2 dif-ferent ecological zones: the closed canopy Ankobagallery forest and the xerophytic spiny forest parcel,3 km from one another (Fig. 1). At Berenty, thegallery forest is a secondary forest, which containsboth native and introduced plant species. Along withnative Tamarindus indica trees, this habitat containsintroduced Pithecellobium dulce, Azadirachta indicaand Leucaena leucocephala (Soma 2006). The forestborders the Mandare River, which provides amplegroundwater for large fruit trees. The river also pro-vides drinking water for L. catta groups, even in themiddle of the dry season when small puddles remainin the large riverbed. One of the reserve’s main waterwells is located in the middle of the forest along withsmall water troughs, which provide easily accessibledrinking water for L. catta groups.

261

Fig. 1. Location of the study site in Madagascar (left) and within Berenty Reserve (right), with Mandare River forming thenorthern boundary of the reserve. The spatial proximity between the habitats of the 2 study groups (gallery forest and spiny

forest) is shown; the remainder of the park is anthropogenic sisal plantation

Endang Species Res 14: 259–270, 2011

The entry point to the spiny forest at BerentyReserve is marked by a nearly 7 m vertical hillabove a scrub forest habitat, which was most likelythe outer edge of the Mandare River in the past,and thus represents the end of the supply ofgroundwater that supports large fruiting trees(Jolly et al. 2006). No extensive studies of spiny for-est ecology at Berenty Reserve currently exist, butall plant species found in the spiny forest areendemic to this specific forest type, which hasevolved to survive in areas with low-nutrient soils,high temperatures during the summers and lowamounts of rainfall (Grubb 2003, Fenn 2003). Thisforest type is dominated by the plant family Di -diereaceae, especially the genus Alluaudia, whichis characterized by tall yet slender trees withthorny trunks and branches, and Euphorbia cande-labra, characterized by low-lying succulent vines(Fenn 2003, Grubb 2003). Gyrocarpus americanus,Agave sisalana and Opuntia vulgaris are also foundin the Berenty spiny forest parcel (Charrier et al.2007). All trees have small and exposed canopiesand are significantly shorter than trees in thegallery forest at Berenty. Compared with thegallery forests, all spiny forest plants are highlyexposed to solar rays, resulting in a lack of a truecanopy and very little shade (Gautier & Goodman2003). Beginning in late October 2006, the spinyforest group periodically moved into the surround-ing sisal plantations to feed on the flower buds,flowers and pith of Agave sisalana. These plantsproduce no canopy or shade and are characterizedby tall exposed stalks with flowers at the tip.

Data collection methods

Data were collected using 15-min continuous timefocal animal sampling methods (Altmann 1974). Allbehaviours of the focal animal were recorded and theexact start and stop time of each behaviour wasnoted. Ad libitum sampling was used to collectunusual occurrences, such as predation attempts onLemur catta and inter-group conflicts between L.catta groups. Time, to the second, was recorded atthe beginning of each behaviour. A total of 94215 min focal animal data sessions were collected(mean no. of sessions: spiny = 54, gallery = 84),totalling 235 h of data collection (Table 1). Data werecollected 6 d per week for the entire study period.Groups were observed on alternating days to ensurethat hours of data collected were similar betweengroups and that changes in flora in either habitat

were accounted for. Within study groups, the daywas separated into 5 periods (morning, 7:00−10:00 h;late morning, 10:00−12:00 h; early afternoon, 12:00−14:00 h; mid-afternoon, 14:00-16:00 h; late afternoon,after 16:00 h). These time periods were used to estab-lish daily patterns of activity within and betweengroups and are referred to numerically as time peri-ods 1 to 5 in the ‘Results’ section.

Behaviours scored in time allocation data were:feeding, foraging, locomotion, sitting, resting, socialbehaviour and vigilance. Resting and sitting werecategorized as inactive behaviours, whereas all otherbehaviours were categorized as active behaviours.Nearest neighbour distance was scored as near(<2 m), medium (2 to 5 m) and far (>5 m).

The total rate of agonism was calculated by divid-ing the total number of agonistic behaviours by thetotal number of hours recorded. The rate of agonismin behavioural contexts was calculated by dividingthe total number of agonism events during thebehaviour by the total duration of the behaviour.

Sample sizes for each group were too small tomeet the assumptions of parametric statistical tests,so comparisons of behaviours between groupswere assessed using non-parametric statisticaltests. The Mann-Whitney U-test was used to com-pare data between groups. Friedman’s ANOVAwas used to test for variations in activity betweendifferent time periods throughout the day withinthe spiny forest group. Results were deemed sig-nificant at p < 0.05.

262

Individual Sex Hours Sessions

Spiny forestLOL F 14.08 56LUC F 13.43 53LIN F 14.03 56DRA M 12 48DAN M 14.16 56DNT M 14.26 57Mean 13.66 54.33

Gallery forestKRS F 18.9 76MAJ F 22.4 90HLN F 21.8 87FNA F 22.1 88GAS M 21.2 85JAC M 20.7 83SEB M 19.9 80

Mean 21.02 84.14

Table 1. Number of hours and sessions of data collected ineach habitat on individual focal Lemur catta. Abbreviations

present names of study individuals. F: female; M: male

Ellwanger et al.: Behavioural flexibility in Lemur catta

RESULTS

Time allocation

Individuals living in the gallery forest group (n = 7)spent a significantly greater proportion of time feed-ing and foraging than those in the spiny forest group(n = 6) (Table 2). Resting occurred significantly moreoften in the spiny forest group than in the gallery for-est group, as did sitting while awake. Lemurs in thegallery forest group engaged in significantly moresocial behaviour compared with the spiny forestgroup (Table 2). There were no significant differ-ences in the proportion of time spent in locomotion.Cumulatively, the gallery forest group spent agreater proportion of time engaged in active behav-iours compared with the gallery forest group.

To identify patterns of behaviour throughout theday, the percentage of time spent in active behav-iours was analyzed within 5 separate time cate-gories. Individuals in the gallery forest group spentsignificantly more time in active behaviours in theearly morning, late morning, and mid-afternooncompared with individuals in the spiny forestgroup, whereas the groups spent similar propor-tions of time in active behaviours in the early andlate afternoon (Table 3). There were significant dif-ferences in the proportion of active behaviours

between all time categories (Friedman test: p <0.001, n = 6) among the spiny forest group only.Post hoc tests were conducted to determine whichtime categories were significantly different fromeach other. Based on an alpha value of 0.001, pair-wise ranked sums were used to compare the pro-portion of active behaviour between time cate-gories. There were no significant differences in thepercentage of active behaviour between time peri-ods 2, 3 and 4 or between time periods 1 and 5 (p =0.001). However, Lemur catta in the spiny forestwere significantly more active in the early morningand later afternoon than during the middle of theday (p = 0.01, Bonferonni adjustment; Table 2).This analysis shows that the spiny forest group lim-ited active behaviours except during the earliestand latest periods of the day.

Feeding and foraging behaviour

The spiny forest group consumed 14 plant speciesduring the study period. Two of these, Agavesisalana (sisal) and Opuntia vulgaris (Mexicanprickly pear), are not native to Madagascar’s spinyforest ecosystem and are found in some semi-deserthabitats outside Madagascar (Felger 2000). Thegallery forest group consumed a total of 11 species,the majority of which (6 out of 11) are not endemic toMadagascar. Both study groups were observed con-suming fruit and leaves of Tamarindus indica andleaves of Opuntia vulgaris. T. indica are locatedthroughout the gallery and scrub forests at Berentybut not in the spiny forest; the spiny forest groupoccasionally moved into the scrub forest area to feedon T. indica.

Individuals in both groups consumed a variety ofplant parts, including mature and immature fruitsand leaves, stems, leaf buds, flowers and leaves ofsucculent plants. Soil was consumed occasionally.The 2 groups differed significantly in their consump-tion of both fruits and flowers. The spiny forest groupspent significantly more time eating fruit (57.1%)compared with the gallery forest group (26.2%) (U =0, p < 0.005), whereas the gallery forest group spentsignificantly more time consuming leaves (U = 0, p <0.005; Fig. 2). The 2 groups showed significant differ-ences in spacing while feeding on the 2 most com-mon food items: fruit and leaves. While feeding onfruit, the spiny forest group spent a significantlygreater proportion of time within 2 m of their nearestneighbour (U = 4, p < 0.05) and a significantly smallerproportion of their feeding time more than 5 m from

263

Behaviour Spiny Gallery U p

Foraging/feeding 14.9 (5) 30 (4.4) 0 <0.005Locomotion 7.6 (1.4) 7.1 (2.3) 17 0.63Resting 52.3 (3.7) 34.3 (5.4) 0 <0.005Sitting 20.5 (4.6) 14.8 (1.3) 0 <0.005Socialising 2.6 (2.9) 8.8 (4.25) 3 <0.01Active 26.4 (7.7) 49.1 (3.3) 0 <0.005Inactive 73.6 (7.6) 50.9 (3.3) 0 <0.005

Table 2. Lemur catta. Median (interquartile range) percent-age of time allocated by lemurs to behaviours between

spiny forest and gallery forest groups

Time period (h) Spiny Gallery U p

7:00−10:00 48 (15.3) 80 (13) 3 <0.00510:00−12:00 15 (8.3) 37 (6.5) 3 <0.00512:00−14:00 8 (9.3) 12 (6) 31.5 0.1414:00−16:00 23 (8.8) 47 (13) 3 <0.005After 16:00 68 (18.5) 73 (4.5) 27 0.45

Table 3. Lemur catta. Median (interquartile range) percent-age of time spent by lemurs in active behaviours betweenspiny forest and gallery forest groups in different time periods

Endang Species Res 14: 259–270, 2011

a nearest neighbour compared with the gallery forestgroup (U = 3, p < 0.05; Fig. 3). The 2 groups also dif-fered in spacing patterns while feeding on leaves.Individuals in the spiny forest group spent a signifi-cantly smaller proportion of their time in close prox-imity to a nearest neighbour (U = 7, p < 0.05) and asignificantly greater proportion of their time far froma nearest neighbour compared with the gallery forestgroup (U = 6, p < 0.05; Fig. 3).

The rate of total agonism was used as a baseline forcomparison of feeding and social agonism. Therewere no significant differences in the rate of totalagonism or social agonism between groups, but indi-viduals in the spiny forest group engaged in a signif-icantly higher rate of agonistic events per feedinghour compared with individuals in the gallery forestgroup (U = 4.5, p < 0.05; Fig. 4). Because of small sam-ple sizes, the rate of feeding agonism in the spiny for-est group was much greater than any other agonismrate in either group.

DISCUSSION

Primates are renowned for their ability successfullyto modify behavioural strategies to cope with envi-ronmental heterogeneity (Hill et al. 2003, Brockman& van Schaik 2005, Jones 2005). Thus, the investiga-tion of activity patterns is central to the study of vary-ing behavioural strategies in space and time. To sus-tain thermoregulation, physiological maintenanceand reproductive capabilities, individuals are likelyto modify activity patterns to increase energy intakeand decrease energy expenditure (Whiten et al.1987, Barton et al. 1992, Dunbar 1992, Menon &Poirier 1996). Species seek to balance energy expen-diture and intake during seasonal periods of resourcescarcity (Overdorff et al. 1997, Remis 1997, Heming-way & Bynum 2005) by using energy maximization orminimization strategies (Strier 1992). Studies ofbehavioural patterns of primates provide uniqueinsights into adaptive strategies when environmentalconditions challenge physiological processes. Fur-

264

Food item

Tim

e sp

ent

feed

ing

(%)

Fruit

Leav

es*

Flower

s*

Leaf

bud

s

Vine/s

tem

Succu

lent

0

20

40

60

80

100

Spiny forest group

Gallery forest group

Fig. 2. Lemur catta. Proportion of time spent feeding on dif-ferent plant parts. Asterisks indicate significant differences

between spiny forest and gallery forest groups at p < 0.05

Nearest neighbour distance

Tim

e sp

ent

feed

ing

(%)

0–2 m*† 2–5 m >5 m*† 0

20

40

60

80

100 Spiny fruitGallery fruitSpiny leavesGallery leaves

Fig. 3. Lemur catta. Proportion of time spent feeding on fruitand leaves at locations with different nearest neighbour distances. Asterisks and daggers indicate significant dif -ferences between spiny forest and gallery forest groups for

fruit and leaves, respectively, at p < 0.05

Agonism type

Ago

nism

eve

nts

h–1

Total Feeding* Social0

2

4

6

8

10Spiny forest group

Gallery forest group

Fig. 4. Lemur catta. Rate of agonistic behaviour by lemurs.Asterisks indicate significant differences between spiny

forest and gallery forest groups at p < 0.05

Ellwanger et al.: Behavioural flexibility in Lemur catta

thermore, studies of resource use and social re -sponses to locally available resources are essential toproper conservation action plans.

Although sample sizes are necessarily limited forthis Near Threatened and declining species, ourstudy suggests that Lemur catta adjust behaviouralstrategies between habitat types during the hot dryseason. The modification of activity patterns is prob-ably associated with ambient temperature and wateravailability, which we propose are limiting factors inthe proportion of time spent in active behaviours byL. catta living in the spiny forest. L. catta occupyingspiny forest habitats spent significantly more timeconsuming fruits, a highly spatially clumped andcontested resource among L. catta (Sauther 1993).These factors produced a much higher rate of feed-ing agonism within the spiny forest group, raisingintriguing questions about resource availability andquality, and their interconnected effects on L. cattafeeding agonism in non-gallery forest habitats.

This study, limited to one season, indicates that L.catta groups can survive, reproduce and lactate inthis harsh environment; however, more data areneeded to understand better the behavioural andecological parameters of L. catta in spiny foreststhroughout southwestern Madagascar (Gould 2006).Below, we discuss potential causes for modificationsin behavioural patterns between L. catta groups indifferent habitats and provide questions that futurestudies of L. catta occupying spiny forest habitatsshould address.

Time allocation

The data from our study show that individualsliving in the spiny forest habitat significantlyreduced their time spent in active behaviours,including feeding, during the dry season. The datasupport the contention that Lemur catta are energyminimizers (Rasamimanana et al. 2006) and thatincreased levels of resting are associated with lesstime spent feeding (Sussman 1977). Both studygroups tended towards bimodal activity patterns,with early morning and late afternoon active peri-ods broken up by a long period of mid-day resting.Significant differences in intergroup activity pat-terns were primarily related to the spiny forestgroup’s extended periods of rest during the intenseheat of mid-day. This group compensated for defi-ciencies in mid-day foraging effort by using abbre-viated yet intensive feeding periods during theearly and later periods of the day.

Energy conservation strategies, which enableprimates to maintain physiological processes whenresource availability is low (Iwamoto & Dunbar1983, Barton et al. 1992, Dasilva 1992, Wright1999, Di Fiore & Rodman 2001) are widespread inprimates (lemurs: Richard 1978, Wright 1999,Norscia et al. 2006; atelines: Milton 1980, Estradaet al. 1999, Di Fiore & Rodman 2001; colobines:Dasilva 1992, Zhou et al. 2007). Lemur catta in thespiny forest may be able to use an energy-mini-mizing strategy by taking advantage of their inher-ently low basal metabolic rate (BMR). L. catta de -viates further from the Kleiber regression line ofprimate BMR to body size than any other primatesampled, including other small-bodied strepsir-rhines (Snodgrass et al. 2007). A low BMR slowsthe rate of oxidation, which in turn lowers theamount of energy processed by the body. All L.catta use energy- minimizing behavioural strategiesbecause of this inherent physiological characteristic(Rasamimanana et al. 2006), but differences inresource availability between habitats likely leadsto variant behavioural patterns. The spiny forestgroup likely maximizes energy retention by in -creasing time spent resting, particularly duringtimes of high ambient temperatures and in spaceswith little shade. Gould et al. (2011) found that L.catta select foods with relatively high levels of pro-tein, calories and water, allowing them to compen-sate for time lost feeding.

Water availability is most likely a limiting factor inthe allotment of active behaviour. Pre-existingreports of water content show that floral families inthe spiny forests, predominantly Didiereaceae, Her-nandiaceae and Euphorbiaceae, have evolved towithstand long periods of drought and high amountsof solar radiation (Fenn 2003) and, therefore, main-tain lower levels of water per part than floral speciesin gallery forests (Grubb 2003). Water content inleaves of Tamarindus indica, the only food speciesshared by both groups, is positively correlated withproximity to the Mandare River (Mertl-Millhollen etal. 2006). Thus, leaves and presumably fruits fromthe spiny forest likely provide significantly less waterneeded for energy production, although the lemursdid benefit by consuming plants such as Opuntiacacti. Furthermore, the 5 plants most commonly con-sumed by the lemurs in the spiny forest containedsignificant amounts of water compared with less fre-quently consumed plant species (Gould et al. 2011).Moreover, individuals in the gallery forest had accessto drinking water from small pools in the MandareRiver and cement troughs in the Ankoba gallery par-

265

Endang Species Res 14: 259–270, 2011

cel. Thus, although groups in the spiny forest canmaintain sufficient hydration by consuming plantparts (Goodman & Langrand 1996, Gould et al. 2011),they may lose more water than the gallery forestgroup because of lack of access to actual standingwater, increased ambient temperatures in the spinyforest, and greater solar exposure due to lack of for-est canopy. Reduced water intake by spiny forestdwelling Lemur catta (compared with gallery forestdwelling conspecifics) particularly in the hot season,could have indirect effects on female reproduction,but reduced activity may be a compensatory strategy.Our study occurred during the birth and lactationperiod, when female L. catta must produce a largequantity of dilute milk to aid in survival of rapidlygrowing infants that suckle on demand (Power et al.2006). A relationship between milk composition andinfant thermoregulation in primates has been sug-gested (Tilden & Oftedal 1997) and maternal produc-tion of dilute milk can be a reproductive survivalstrategy, particularly when infants must cope withhigh temperatures, such as in the spiny forest (Gouldet al. 2011). Therefore, although no sex differences inresting time were found, lactating females could stillbenefit from engaging in such energy-minimizingbehaviours (Rasamimanana et al. 2006) in a spiny for-est habitat, because reduced activity would result inless water loss.

High ambient temperatures have been consideredto have a significant influence on primate activitycycles (Macaca nemestrina, Bernstein 1972; Papiocynocephalus cynocephalus, Stelzner 1988, Pochron2000; Papio cynocephalus ursinus, Hill 2006) be -cause evaporative water loss in primates increaseswhen ambient temperature rises over 30°C andrises sharply when temperatures exceed 40°C (Eli-zondo 1977). Stelzner (1988) has suggested thatambient temperature is a primary constraint to be -haviour, as primate physiological abilities to reducethermal stress are limited. Our study was conductedduring some of the hottest months in southernMadagascar, with daily maximum temperatures inthe Berenty spiny forest and surrounding matrix attimes reaching over 40°C; in December and Janu-ary, temperatures would rise even further in theentire region (Jolly et al. 2006). The spiny forestgroup, therefore, most likely increased time spentresting in the little shade that exists in this habitatto avoid intense solar radiation and to decreaseevaporative water loss. Energy-intensive behav-iours, especially locomotion and foraging, tookplace at a fast pace and were most pronouncedwhen temperatures were at their lowest.

Foraging strategy

Lemur catta quickly modify their feeding strategiesas resources become available in gallery forest habi-tats (Sauther 1994, 1998, Soma 2006). The presentstudy shows that L. catta alter foraging strategies be-tween habitat types based on the available flora spe-cies within their home ranges. Furthermore, bothgroups fit into the dietary niche of flexible folivore/fru-givore previously ascribed to L. catta (Jolly 1966,Sussman 1977, Sauther 1994, Sauther et al. 1999,Soma 2006, Gould 2006). However, the difference inselection between plant reproductive parts and leavesbetween groups was striking. L. catta in the spiny for-est focused their foraging effort on locating smallfruits, especially from Gyrocarpus americanus andTarenna sp. and consuming leaves of succulent spe-cies such as Alluaudia procera, Aloe vahome and Op-untia vulgaris (Gould et al. 2011). Furthermore, to-wards the conclusion of the study, the spiny forestgroup spent much of its foraging time consumingflowers of Agave sisalana in fields outside of the spinyforest parcel. In contrast, L. catta in the gallery forestspent much of their foraging time consuming leaves,especially from Leucaena lercoephalla, Pithecel-lobium dulce and Azadirachta indicus, and consumedfruits from P. dulce and Tamarindus indica as they be-came available. A comparison of nutrient intake be-tween populations living in the 2 diverse habitatswould be useful to understand whether activity-mini-mization strategies are associated with significantlylower levels of energy and nutrient intake.

Distinctly different patterns of social spacing whileforaging between the 2 groups are most likely associ-ated with significant differences in Lemur catta forag-ing agonism. Contest competition is more likely to oc-cur when primate groups feed on plant reproductiveparts, which are often spatially clumped, comparedwith leaves, which tend to be ubiquitous throughouthabitats (Wrangham 1980, van Schaik 1989). We sug-gest that because fruit is clumped in small treecrowns in the spiny forest, L. catta in that habitat en-gaged in higher levels of feeding competition whilesearching for these preferred food items. During thestudy period, L. catta in the gallery forest increasedinter-individual spacing while consuming fruit be-cause of increased spatial distances between fruititems in large Pithecellobium dulce and Tamarindusindica trees, and decreased individual spacing whileconsuming leaves, which are more plentiful and spa-tially clumped. Further research on L. catta feedingbehaviour in spiny forest habitats can further identifypatterns in relation to food type, seasonal effects on

266

Ellwanger et al.: Behavioural flexibility in Lemur catta

feeding competition, competition over specific plantpart species, and the possible effect of feeding com-petition on differently ranked individuals.

CONCLUSIONS

We suggest that because Lemur catta successfullynegotiate the ecological challenges of the spiny for-est in Berenty Reserve, other populations of this spe-cies could consistently occupy spiny forest habitats ifenough of the important food resources were avail-able. Unfortunately, very little is known about theseasonal and annual patterns of resource use, distri-bution and activity of L. catta populations living inother spiny forests (although at the time of writing,one year-long study was underway in a differentspiny forest region of southern Madagascar; M.LaFleur unpubl. data). Although population densityis relatively low in non-gallery forest habitats (Suss-man et al. 2003), these spiny forest groups are essen-tial in preserving a diverse population gene pool andretaining the possibility of conserving the speciesoutside of a small number of gallery forests nearrivers. Goodman et al. (2006) have noted that ring-tailed lemurs may disperse through low-resourcecorridors; thus maintaining primary or secondary for-est throughout southern Madagascar may be vital tothis species’ future. Moreover, as seed dispersers, L.catta are vital for the survival of both gallery forests(Blumenfeld-Jones et al. 2006) and spiny forests(Bodin et al. 2006). The endemic spiny forests arevanishing at a rapid pace because of anthropogenicdeforestation, and the remaining small fragments ofthis endemic forest type should be viewed as essen-tial to maintaining a genetically variable primatespecies (Bodin et al. 2006). Thus, in the evaluation ofhabitats to be designated as protected for L. catta, allhabitat types that contain the species, especiallyxerophytic spiny forest, should be considered.

Acknowledgements. We thank the de Heaulme family andthe Département des Eaux et Forêts, Madagascar, for grant-ing us permission to work at Berenty Reserve. Funding forthis project was provided by a Natural Sciences and Engi-neering Research Council of Canada (NSERC) to L.G. Allresearch described in this manuscript complied with proto-cols approved by the University of Victoria’s Animal CareCommittee.

LITERATURE CITED

Altmann J (1974) Observational study of behaviour: sam-pling methods. Behaviour 49: 227−267

Barton R, Whiten A, Strum S, Bryne R, Simpson A (1992)Habitat use and resource availability in baboons. AnimBehav 43: 831−844

Bernstein I (1972) Daily activity cycles and weather influ-ences on a pigtail monkey group. Folia Primatol 18: 390−415

Blumenfeld-Jones K, Randriamboavonjy T, Williams G,Mertl-Millhollen A, Pinkus S, Rasamimanana H (2006)Tamarind recruitment and long-term stability in thegallery forest at Berenty, Madagascar. In: Jolly A, Suss-man R, Koyama N, Rasamimanana H (eds) Ring-tailedlemur biology: Lemur catta in Madagascar. Springer,New York, NY, p 69−85

Bodin O, Tengo M, Norman A, Lundberg J, Elmqvist T(2006) The value of small size: loss of forest patches andecological thresholds in southern Madagascar. Ecol Appl16: 440−451

Brockman DK, van Schaik CP (2005) Seasonality in pri-mates: studies of living and extinct human and non-human primates. Cambridge University Press, NewYork, NY

Casse T, Milhoj A, Ranaivoson S, Randriamanarivo J (2004)Causes of deforestation in southwestern Madagascar: What do we know? For Policy Econ 6: 33−48

Charrier A, Hladik A, Simmen B (2007) Stratégie alimen-taire et dominance des femelles Propithèque de Ver-reauxi (Propithecus v. verreauxi) dans la forêt àDidiereaceae du sud de Madagascar. Rev Ecol Terre Vie62: 257−263

Dasilva G (1992) The western black-and-white colobus as alow-energy strategist: activity budgets, energy expendi-ture and energy intake. J Anim Ecol 61: 79−91

Di Fiore A, Rodman P (2001) Time allocation patterns in low-land wooly monkeys (Lagothrix lagotricha poeppigii) in aneotropical terra firma forest. Int J Primatol 22: 449−480

Dunbar R (1992) Time: a hidden constraint on the behav-ioural ecology of baboons. Behav Ecol Socioecol 31: 35−49

Elizondo R (1977) Temperature regulation in primates: In: Robertshaw A (ed) International review of physiology,Vol. 5. University Park Press, Baltimore, MD

Estrada A, Juan-Solano S, Martinez T, Coates-Estrada R(1999) Feeding and general activity patterns of ahowler monkey (Alouatta palliata) troop living in a forest fragment at Los Tuxtlas, Mexico. Am J Primatol48: 167−183

Felger R (2000) Flora of the Gran Desierto and Rio Coloradoof northwestern Mexico. University of Arizona Press,Tucson, AZ

Fenn M (2003) The spiny forest ecoregion. In: Goodman S,Benstead J (eds) The natural history of Madagascar. Uni-versity of Chicago Press, Chicago, IL

Fenn M, Rebara F (2003) Present migration tendencies andtheir impacts in Madagascar spiny forest ecoregion.Nomad People 7: 123−137

Ganzhorn J, Fietz J, Rakatovao E, Schwab D, Zinner D(1999) Lemurs and the regeneration of dry deciduous for-est in Madagascar. Conserv Biol 13: 794−804

Gautier L, Goodman SM (2003) Introduction to the flora ofMadagascar. In Goodman S, Benstead J (eds) The nat-ural history of Madagascar. University of Chicago Press,Chicago, IL, p 229−256

Goodman S, Langrand O (1996) A high mountain populationof the ring-tailed lemur (Lemur catta) on the AndringitraMassif, Madagascar. Oryx 30: 259−268

267

Endang Species Res 14: 259–270, 2011

Goodman S, Rasolonandrasana B (2001) Elevational zona-tion of birds, insectivores, rodents, and primates on theslopes of the Andringitra Massif, Madagascar. J Nat Hist3(2): 285–305

Goodman S, Rakotoarisoa S, Wilme L (2006) The distributionand biogeography of the ring-tailed lemur (Lemur catta).In: Jolly A, Sussman R, Koyama N, Rasamimanana H(eds) Ring-tailed lemur biology: Lemur catta in Mada-gascar. Springer, New York, NY, p 3−15

Gould L (1990) The social development of free-ranginginfant Lemur catta at Berenty Reserve, Madagascar. Int JPrimatol 11: 297−318

Gould L (1992) Alloparenting care in free-ranging Lemurcatta at Berenty Reserve, Madagascar. Folia Primatol 58: 72−83

Gould L (1996a) Vigilance behaviour during the birth andlactation season in naturally occurring ring-tailedlemurs. Int J Primatol 17: 331−347

Gould L (1996b) Male-female affiliative relationships in nat-urally occurring ring-tailed lemurs (Lemur catta) at theBeza-Mahafaly Reserve, Madagascar. Am J Primatol 39: 63−78

Gould L (1997a) Affiliative relationships between adultmales and immature group members in naturally occur-ring ring-tailed lemurs (Lemur catta). Am J Primatol 103: 163−171

Gould L (1997b) Intermale affiliative behaviour in ring-tailed lemurs (Lemur catta) at the Beza-MahafalyReserve, Madagascar. Primates 38: 15−30

Gould L (2006) Lemur catta ecology: what we know andwhat we need to know. In: Gould L, Sauther M (eds)Lemurs: ecology and adaptation. Springer, New York,NY, p 255−274

Gould L, Sussman R, Sauther M (1999) Natural disasters andprimate populations: The effects of a 2-year drought on anaturally occurring population of ring-tailed lemurs(Lemur catta) in southwestern Madagascar. Int J Prima-tol 20: 69−84

Gould L, Sussman R, Sauther M (2003) Demographic andlife history patterns in a population of ring-tailed lemurs(Lemur catta) at Beza Mahafaly Reserve, Madagascar: a15-year perspective. Am J Phys Anthropol 120: 182−194

Gould L, Constabel P, Mellway R, Rambeloarivony H (2009)Condensed tannin intake in spiny-forest-dwelling Lemurcatta at Berenty Reserve, Madagascar, during reproduc-tive periods. Folia Primatol 80: 249−263

Gould L, Cameron A, Gabriel D (2010) Lemurs on the rocks: preliminary study of ring-tailed lemur demography,habitat use, and feeding ecology in rocky outcrop habitatin south-central Madagascar. Am J Phys AnthropolSuppl 50: 114

Gould L, Power ML, Ellwanger N, Rambeloarivony H (2011)Feeding behaviour and nutrient intake in spiny forest-dwelling ring-tailed lemurs (Lemur catta) during earlygestation and early to mid-lactation periods: compensat-ing in a harsh environment. Am J Phys Anthropol 145: 469−479

Green GM, Sussman RW (1990) Deforestation history of theeastern rainforests of Madagascar from satellite images.Science 248: 212−215

Grubb P (2003) Interpreting some outstanding features ofthe flora and vegetation of Madagascar. Perspect PlantEcol Evol Syst 6: 125−146

Hemingway C, Bynum N (2005) The influence of seasonalityon primate diet and ranging. In: Brockman D, van Schaik

C (eds) Seasonality in primates: studies of living andextinct human and non-human primates. CambridgeUniversity Press, New York, NY, p 57−104

Hill RA (2006) Thermal constraints on activity schedulingand habitat choice in baboons. Am J Phys Anthropol 129: 242−249

Hill RA, Barrett L, Gaynor D, Ingrill T, Dixon P, Payne H,Henzi SP (2003) Day length, latitude and behavioural(in)flexibility in baboons (Papio cynocephalus ursinus).Behav Ecol Sociobiol 53: 278−286

IUCN (2009) IUCN Red List of Threatened Species. Version2009.2. www.iucnredlist.org

Iwamoto T, Dunbar R (1983) Thermoregulation, habitatquality, and the behavioural ecology of gelada baboons.J Anim Ecol 52: 357−366

Jolly A (1966) Lemur behaviour: a Madagascar field study.University of Chicago Press, Chicago, IL

Jolly A (1984) The puzzle of female feeding priority. In: Small M (ed) Female primates: studies by women prima-tologists. Alan R. Liss, New York, NY, p 179–215

Jolly A, Dobson A, Rasamimanana H, Walker J, O’Connor S,Solberg M, Perel V (2002) Demography of Lemur catta atBerenty Reserve, Madagascar: effects of troop size, habi-tat and rainfall. Int J Primatol 23: 327−353

Jolly A, Koyama N, Rasamimanana H, Crowley H, WilliamsG (2006) Berenty Reserve: a research site in southernMadagascar. In: Jolly A, Sussman R, Koyama N, Rasami-manana H (eds) Ring-tailed lemur biology: Lemur cattain Madagascar. Springer Press, New York, NY, p 32−42

Jones C (2005) Behavioural flexibility in primates: causesand consequences. Springer, Chicago, IL

Kelley EA, Sussman RW, Muldoon KM (2007) The status oflemur species at Antserananomby: an update. PrimateConserv 22: 71−77

Koyama N, Nakamichi M, Oda R, Ichino S, Miyamoto N,Takahata Y (2001) A 10-year summary of reproductiveparameters for ring-tailed lemurs at Berenty, Madagas-car. Primates 42: 1−14

Menon S, Poirier F (1996) Lion-tailed macaques (Macacasilenus) in a disturbed forest fragment: activity patternsand time budgets. Int J Primatol 17: 969−985

Mertl-Millhollen AS, Rambeloarivony H, Miles W, KaiserVA and others (2006) The influence of tamarind treequality and quantity on Lemur catta behaviour. In: JollyA, Sussman R, Koyama N, Rasamimanana H (eds) Ring-tailed lemur biology: Lemur catta in Madagascar.Springer, New York, NY, p 102−118

Milton K (1980) The foraging strategy of howling monkeys: a study in primate economics. Columbia UniversityPress, New York, NY

Myers N, Mittermeier R, Mittermeier C, Gustavo A, Kent J(2000) Biodiversity hotspots for conservation priorities.Nature 403: 853−858

Nakamichi M, Koyama N (1997) Social relationships amongring-tailed lemurs (Lemur catta) in two free-rangingtroops at Berenty Reserve, Madagascar. Int J Primatol 18: 73−93

Norscia I, Carrai V, Borgognini-Tarli S (2006) Influence ofdry season and food quality on behaviour and feedingstrategy of Propithecus verreauxi in Kirindy, Madagas-car. Int J Primatol 27: 1001−1022

Overdorff DJ, Strait SG, Telo A (1997) Seasonal variation inactivity and diet in a small bodied folivorous primate,Hapalemur griseus, in southeastern Madagascar. Am JPrimatol 43: 211−223

268

Ellwanger et al.: Behavioural flexibility in Lemur catta 269

Pereira M, Strohecker R, Cavigelli S, Hughes C, Pearson D(1999) Metabolic strategy and social behaviour inLemuiridae. In: Rakotosamimanana B, Rasamimanana H,Ganzhorn J, Goodman S (eds) New directions in lemurstudies. Kluwer Academic, New York, NY, p 93–118

Pochron S (2000) Sun avoidance in yellow baboons (Papiocynocephalus cynocephalus) of Ruaha Park, Tanzania: variations with season, behaviour, and weather. Int JMeteorol 44: 141−147

Power M, Tilden C, Oftedal O (2006) Patterns of milk com-position in prosimian primates. Am J Primatol 68(Suppl1): 131

Pride E (2003) The socio-endocrinology of group size inLemur catta. PhD dissertation, Princeton University,Princeton, NJ

Rasamimanana H, Rafidinarivo E (1993) Feeding behaviourof Lemur catta females in relation to their physiologicalstate. In: Kappeler P, Ganzhorn J (eds) Lemur social sys-tems and their ecological basis. Plenum Press, New York,NY, p 271–295

Rasamimanana H, Andrianome V, Ramibeloarivony H, Pas-quet P (2006) Male and female ring-tailed lemurs’ ener-getic strategy does not explain female dominance. In: Jolly A, Sussman R, Koyama N, Rasimimanana H (eds)Ring-tailed lemur biology: Lemur catta in Madagascar.Springer, New York, NY, p 271−295

Remis E (1997) Western lowland gorillas (Gorilla gorillagorilla) as seasonal frugivores: use of variable resources.Am J Primatol 43: 87−109

Richard A (1978) Behavioral variation: case study of Mala-gasy lemur. Associated University Press, Cranburry, NJ

Sauther ML (1991) Reproductive behaviour of free-rangingLemur catta at Beza Mahafaly Special Reserve, Mada-gascar. Am J Phys Anthropol 84:463–477

Sauther ML (1993) Resource competition in wild populationsof ring-tailed lemurs (Lemur catta): implications forfemale dominance. In: Kappeler P, Ganzhorn J (eds)Lemur social systems and their ecological basis. PlenumPress, New York, NY, p 135–152

Sauther ML (1994) Wild plant use by pregnant and lactatingring-tailed lemurs, with implications for early hominidforaging. In: Etkin N (ed) Eating on the wild side. Univer-sity of Arizona Press, Tucson, AZ, p 240–256

Sauther ML (1998) Interplay of phenology and reproductionin ring-tailed lemurs: implications for ring-tailed lemurconservation. Folia Primatol 69: 309−320

Sauther ML, Sussman RW, Gould L (1999) The socioecologyof the ring-tailed lemur: thirty-five years of research.Evol Anthropol 8: 120−132

Scott D, Brown D, Hahood S, Denton B, Silburn A, Rakaton-draparany F (2006) The impacts of forest clearance onlizards, small mammals and bird communities in the aridspiny forest, southern Madagascar. Biol Conserv 127: 72−87

Snodgrass J, Leonard W, Robertson M (2007) Primate bio-metrics: an evolutionary perspective. In: Ravosa M,Dagosto M (eds) Primate origins: adaptations and evolu-tion. Springer, New York, NY, p 703–737

Soma T 2006. Tradition and novelity: Lemur catta feedingstrategy on introduced tree species at Berenty Reserve.In: Jolly A, Sussman R, Koyama N, Rasamimanana H (ed)Ring-tailed lemur biology: Lemur catta in Madagascar.Springer Press, New York, NY, p 141−159

Stelzner J (1988) Thermal effects on movement patterns ofyellow baboons. Primates 29: 91−105

Strier KB (1992) Ateline adaptations: behavioral strategiesand ecological constraints. Am J Phys Anthropol 88: 515−524

Sussman RW (1974) Ecological distinctions in overlappingLemur species. In: Martin R, Doyle G, Walker A (eds)Prosimian biology. Duckworth Publishing, London,p 75–108

Sussman RW (1977) Feeding behaviour of Lemur catta andLemur fulvus. In: Clutton-Brock T (ed) Primate ecology: studies of feeding and ranging behaviour in lemurs, mon-keys, and apes. Academic Press, New York, NY p 1–36

Sussman RW (1991) Demography and social organization offree-ranging Lemur catta in the Beza Mahafaly Reserve,Madagascar. Am J Phys Anthropol 84: 43−58

Sussman RW (1992) Male life history and intergroup mobil-ity among ring-tailed lemurs. Int J Primatol 13: 395−413

Sussman RW (2002) Adaptive array of lemurs of Madagascarrevisited. Evol Anthropol 11: 75−78

Sussman RW, Green G, Portman I, Andrianasolondraibe O,Ratsirarson J (2003) A survey of the habitat of Lemurcatta in southwestern and southern Madagascar. PrimateConserv 19: 32−57

Tilden CD, Oftedal OT (1997) Milk composition reflects pat-terns of maternal care in prosimian primates. Am J Pri-matol 41: 195−211

van Schaik C (1989). The ecology of social relationshipsamongst female primates. In: Standen V, Foley R (eds)Comparative socioecology: the behavioural ecology ofhumans and other mammals. Blackwell Scientific Publi-cations, Boston, MA, p 195–218

Wells D (2003) The spiny forest ecoregion. In: Goodman S,Benstead J (eds) A natural history of Madagascar. Uni-versity of Chicago Press, Chicago, IL, p 384–389

Whiten A, Byrne RW, Henzi SP (1987) The behavioural ecol-ogy of mountain baboons. Int J Primatol 8: 367−388

Wrangham RW (1980) An ecological model of female-bonded primate groups. Behaviour 75: 262−300

Wright PC (1999) Lemur traits and Madagascar ecology.Yearb Phys Anthropol 110(Suppl. 29): 31−72

Zhou Q, Ii F, Huang C, Li M, Ren B, Luo B (2007) Seasonalvariation in the activity patterns and time budgets of Tra-chypithecus francoisi in the Nonggang Nature Reserve,China. Int J Primatol 28: 657−671

Endang Species Res 14: 259–270, 2011270

Taxonomic family Species Food type

Spiny forestAgavea Agave sisalana Succulent leavesBurserceae Commiphora sp. Leaf budsBoraginaceae Cordia caffra LeavesCactaceaea,b Opuntia sp. Succulent leavesCaesalpiniodaeb Tamarindus indica Fruit, leaves, leaf budsCapparaceae Maeura folfimus LeavesDidiereaceae Alluadia procera Flowers, leaves, fruitEuphorbia Euphorbia sp. StemsHernandiaceae Gyrocarpus americanus Flowers, leaves, fruitLiliaceae Aloe sp. Succulent leavesMimosoideae Acacia sp. LeavesRibiacae Tarenna sp. FruitSalvadoraceae Azima tetracantha Leaves, flowersSalvadoraceae Salvadora augustifolia Fruit, leaf buds

Soil

Gallery forestCactaceaea,b Opuntia sp. Succulent leavesCaesalpiniodeaeb Tamarindus indica Leaves, leaf buds, fruitCapparaceaea Cordia sp. Leaves and fruitCapparaceae Craveta excelsa FlowersCapparidaceae Cadaba virgata LeavesConvolvulaceaea Iporea robbers FlowersMimosoideae Acacia rovumae LeavesMimosoideaea Leucaena lercoephalla LeavesMimosoideaea Pithecellobium dulce Fruit, flowersMeleaceaea Azadirachta indica Leaves

Soilanon-native species to Madagascar (i.e. introduced and non-endemic);bspecies consumed by both groups

Appendix. Table A1. Plant species eaten by spiny forest and gallery forestgroups of Lemur catta. A number of individuals also selected soil off the

ground as a supplementary part of their diet

Editorial responsibility: Phillip Stephens,Durham, UK

Submitted: May 25, 2010; Accepted: June 14, 2011Proofs received from author(s): September 16, 2011